HomePhabricator

[ARM] Don't pessimize i32 vselect.

Description

[ARM] Don't pessimize i32 vselect.

The underlying issues surrounding codegen for 32-bit vselects have been resolved. The pessimistic costs for 64-bit vselects remain due to the bad
scalarization that is still happening there.

I tested this on A57 in T32, A32 and A64 modes. I saw no regressions, and some improvements.

From my benchmarks, I saw these improvements in A57 (T32)
spec.cpu2000.ref.177_mesa 5.95%
lnt.SingleSource/Benchmarks/Shootout/strcat 12.93%
lnt.MultiSource/Benchmarks/MiBench/telecomm-CRC32/telecomm-CRC32 11.89%

I also measured A57 A32, A53 T32 and A9 T32 and found no performance regressions. I see much bigger wins in third-party benchmarks with this change

Differential Revision: http://reviews.llvm.org/D14743

Details

Committed
chatur01Nov 17 2015, 9:25 AM
Differential Revision
D14743: [ARM] Don't pessimize i32 vselect.
Parents
rL253348: function names start with a lower case letter; NFC
Branches
Unknown
Tags
Unknown